This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.1958
Accuracy: 0.9470

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 64
eval_batch_size: 1
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1042	1.0	21	1.0970	0.3489
1.0711	2.0	42	1.0849	0.3863
1.0265	3.0	63	1.0484	0.4424
0.9343	4.0	84	0.9641	0.5202
0.9037	5.0	105	0.8967	0.5639
0.7794	6.0	126	0.8947	0.5607
0.7517	7.0	147	0.7444	0.6854
0.6897	8.0	168	0.7788	0.6542
0.6639	9.0	189	0.9096	0.5981
0.7325	10.0	210	0.7083	0.6947
0.6112	11.0	231	0.6447	0.7383
0.5403	12.0	252	0.7023	0.6978
0.446	13.0	273	0.7003	0.7009
0.481	14.0	294	0.7621	0.7103
0.4406	15.0	315	0.5552	0.7882
0.3905	16.0	336	0.6945	0.7445
0.3573	17.0	357	0.4493	0.8318
0.4212	18.0	378	0.5443	0.7975
0.3467	19.0	399	0.4746	0.8006
0.275	20.0	420	0.4475	0.8318
0.2847	21.0	441	0.5923	0.7882
0.3065	22.0	462	0.4106	0.8629
0.2062	23.0	483	0.7714	0.7352
0.281	24.0	504	0.3681	0.8816
0.2239	25.0	525	0.3852	0.8411
0.2132	26.0	546	0.5309	0.8037
0.2846	27.0	567	0.4192	0.8349
0.1943	28.0	588	0.5426	0.7913
0.1594	29.0	609	0.4108	0.8380
0.2061	30.0	630	0.3541	0.8692
0.1616	31.0	651	0.2926	0.9097
0.1755	32.0	672	0.4088	0.8442
0.156	33.0	693	0.5390	0.8131
0.3471	34.0	714	0.4996	0.8255
0.1627	35.0	735	0.3407	0.8847
0.1332	36.0	756	0.2970	0.8972
0.2394	37.0	777	0.4211	0.8411
0.1086	38.0	798	0.3548	0.8847
0.1369	39.0	819	0.3874	0.8754
0.1235	40.0	840	0.2755	0.9065
0.132	41.0	861	0.3844	0.8816
0.1109	42.0	882	0.4368	0.8629
0.1291	43.0	903	0.3609	0.8754
0.1071	44.0	924	0.2968	0.9065
0.0967	45.0	945	0.3095	0.8847
0.1031	46.0	966	0.2942	0.9034
0.1124	47.0	987	0.2314	0.9252
0.085	48.0	1008	0.3651	0.8879
0.1031	49.0	1029	0.4515	0.8598
0.112	50.0	1050	0.2458	0.9190
0.11	51.0	1071	0.3609	0.8972
0.0742	52.0	1092	0.3382	0.9065
0.0783	53.0	1113	0.3365	0.9097
0.0748	54.0	1134	0.3213	0.9065
0.0983	55.0	1155	0.3021	0.9034
0.0651	56.0	1176	0.2968	0.9128
0.0762	57.0	1197	0.3651	0.8660
0.0831	58.0	1218	0.3192	0.9003
0.0989	59.0	1239	0.3015	0.8910
0.0936	60.0	1260	0.3160	0.8879
0.0725	61.0	1281	0.2810	0.8972
0.0779	62.0	1302	0.2108	0.9252
0.0565	63.0	1323	0.2214	0.9315
0.1442	64.0	1344	0.2242	0.9221
0.051	65.0	1365	0.2143	0.9128
0.0901	66.0	1386	0.3374	0.8941
0.08	67.0	1407	0.2368	0.9252
0.0425	68.0	1428	0.3098	0.9003
0.0603	69.0	1449	0.2638	0.9097
0.0604	70.0	1470	0.2012	0.9439
0.0405	71.0	1491	0.1795	0.9439
0.0526	72.0	1512	0.4807	0.8505
0.0728	73.0	1533	0.2587	0.9190
0.0406	74.0	1554	0.2363	0.9408
0.063	75.0	1575	0.3414	0.9034
0.0498	76.0	1596	0.2719	0.9159
0.0377	77.0	1617	0.2965	0.9128
0.0573	78.0	1638	0.2341	0.9377
0.096	79.0	1659	0.2503	0.9252
0.0373	80.0	1680	0.3416	0.9003
0.0485	81.0	1701	0.3115	0.9159
0.0502	82.0	1722	0.2318	0.9346
0.0559	83.0	1743	0.2506	0.9097
0.0446	84.0	1764	0.2691	0.9159
0.0344	85.0	1785	0.2695	0.9221
0.0306	86.0	1806	0.2747	0.9097
0.0404	87.0	1827	0.1680	0.9564
0.036	88.0	1848	0.2653	0.9221
0.0364	89.0	1869	0.1936	0.9408
0.0418	90.0	1890	0.2321	0.9346
0.0443	91.0	1911	0.2132	0.9408
0.0409	92.0	1932	0.2601	0.9190
0.0431	93.0	1953	0.2282	0.9439
0.0258	94.0	1974	0.2003	0.9346
0.0294	95.0	1995	0.2033	0.9439
0.0336	96.0	2016	0.1445	0.9595
0.0241	97.0	2037	0.2268	0.9346
0.037	98.0	2058	0.1920	0.9408
0.03	99.0	2079	0.1676	0.9533
0.0227	100.0	2100	0.1958	0.9470

Framework versions

Transformers 4.41.2
Pytorch 2.3.0
Datasets 2.19.2
Tokenizers 0.19.1

talli96123/meat_calssify_fresh_crop_fixed_overlap_epoch100_V_0_1

meat_calssify_fresh_crop_fixed_overlap_epoch100_V_0_1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

talli96123/
meat_calssify_fresh_crop_fixed_overlap_epoch100_V_0_1